I didn't CC, however I would prefer a CC on reply.
Tony Finch wrote: > On Fri, 26 Aug 2005, Wakko Warner wrote:
> >
> > How difficult would it be to detect this and act like tls-on-connect?
>
> It's heavily dependent on network delays and client speed. The server
> would have to delay its initial banner for a few seconds so that it
> doesn't disrupt the SSL negotiation. This would slow down STARTTLS message
> submission and it would be unreliable. Generally a bad idea.
I thought about that as I sent it. I suppose the connection messages would
confuse the tls-on-connect client and wouldn't work any way. I've never
tried that, so I don't know. If it could be done w/o problems it might not
be a bad idea so long as it's always the same init string. I'm thinking of
this in technical terms only (instead of it being good/bad idea).
--
Lab tests show that use of micro$oft causes cancer in lab animals
